Magazine – PDF - Cal Lab Magazine

More documents

Recommendations

Info

An Introduction to Mass Metrology in Vacuum Patrick J. Abbott, Zeina J. Jabbour Figure 4. Diagrams of differential (top) and absolute (bottom) capacitance diaphragm gauges. Px and Pr are the unknown and reference pressures respectively. Cx and Cr are the capacitor plates on the unknown and reference sides [40]. for ISO-100 fittings. Small diameter connecting hardware in effect “chokes” the high vacuum pump, and the effect on pumping speed is analogous to draining a swimming pool with a drinking straw. Conductance can be maximized by fastening the inlet port of the high vacuum pump directly to an equal diameter port on the vacuum chamber. However, this is a risky proposition with mass metrology in that vibrations from the pump may couple into the sensitive mass comparator and introduce a source of instability. While it is good vacuum design practice to keep the pumps as close as possible to the vacuum chamber, vibration isolation measures may need to be considered, such as adding a bellows between the pump and chamber or simply locating the pump a moderate distance away from the chamber. Davidson [9] has eliminated the vibration problem by using a turbomolecular pump having magnetically levitated blades. Using a wide diameter pipe (100 mm is usually sufficient) to locate the pump a few meters from the chamber will reduce the effect of vibrations and heat that are generated by both the primary and high vacuum pumps. The connecting pipe should have as few bends as possible, as each bend reduces the total conductance. The effective conductance between pump and chamber may be Cal Lab: The International Journal of Metrology calculated for a given vacuum system, but the analytical expressions change as a function of gas flow regime. The formulae for making these calculations may be found in any of the vacuum technology texts appearing in the reference section. In most cases, the above sources of gas, along with component conductances and pumping speeds can be calculated, measured, or estimated with sufficient accuracy in order to create an effective design. Measuring Pressure Types of pressure gauges Pressure is force per unit area, and is commonly measured over 16 decades from about 10 9 Pa to 10 -7 Pa . To make accurate measurements over such a wide range requires many different technologies. Pressure gauges may be direct or indirect reading. In direct reading gauges, a non-gas dependent change in some physical property of the gauge occurs in response to the application of pressure. An example of this type is the capacitance diaphragm gauge shown in Figure 4 [40], in which a thin metal diaphragm stretched between two electrodes deflects away from its equilibrium position with the application of pressure; the deflection changes the capacitance between the diaphragm and the electrodes, and this change is converted to a pressure indication by conditioning electronics. Indirect reading gauges respond to some other quantity that is proportional to pressure and frequently dependent on gas species. An example of this is the thermocouple gauge which measures heat flow as a function of pressure [41]. The type of pressure gauge used in a particular application depends on the pressure range of interest, the desired accuracy of the measurement, compatibility with other apparatus, and of course monetary cost. Table 2 below lists some of the more common gauges that are useful for mass in vacuum metrology along with some of their characteristics. Pressure Gauge Calibration Table 2 presents accuracy estimates for uncalibrated gauges, and these estimates can vary widely according to gauge type [42]. Thermal conductivity gauges and ionization gauges may be inaccurate by as much as a factor of two. For proper use, any gauge should be calibrated prior to making measurements. A calibration will provide the user with a pressure-dependent correction factor to apply to the gauge’s indicated pressure value, as well as an uncertainty estimate on the corrected pressure values at the time of calibration. The uncertainty estimate is critical, and the user should not expect better results than the calibration uncertainty dictates. The user must always operate the pressure gauge in a manner consistent with its calibration; failure to do so will void the calibration. Calibration conditions to observe include using the exact same hardware as was used for the calibration (e.g., one can’t mix and match gauge controllers with gauge heads 30 Oct • Nov • Dec 2011
Type of Gauge Pressure Range Direct/ Indirect in general), proper operating parameters, the same gas as the gauge was calibrated with, mounting the gauge in the proper orientation, and observing the safe pressure range of the gauge. Even with proper calibration, the response of many pressure gauges tends to drift away from calibration with time and use. Several factors influence this drift including exposure to atmospheric pressure, accidental over pressurization, rough handling, or inherent limitations of the gauge’s technology. For instance, inexpensive thermocouple gauges may not be stable to within 20% even over the course of the calibration! It is essential to choose a gauge and calibration service that is appropriate for the application. Many vendors offer calibrations that are traceable [43] to the National Institute of Standards and Technology (NIST) or other National Metrology Institutes. Only history and experience can determine how frequently your pressure gauges need to be calibrated. It cannot be overstated that when choosing a calibration service, it is prudent to be mindful of the required uncertainty for the pressure measurement application. For mass-in-vacuum metrology, the calculation above determined that the upper limit of pressure is 0.1 Pa. To measure the difference between 0.1 Pa and 0.09 Pa, it’s necessary to have better than 10% uncertainty. Typically, lower calibration uncertainty translates into higher gauge cost. In this case, paying for a gauge that can hold a 0.1% calibration uncertainty is not necessary when 1% is more than adequate. Gas Species Dependent? Principle Accuracy (uncalibrated) Thermo-couple 1 kPa – 0.01 Pa Indirect Yes Heat Transfer ± 20% Low Convection Enhanced Pirani Capacitance Diaphragm 100 kPa – 0.1 Pa Indirect Yes Heat Transfer ± 20% Low 100 kPa – 10 -3 Pa Direct No Pressure Dependent Capacitance Change Piezoresistive 100 kPa – 10 Pa Direct No Pressure Dependent Change in Resistance Resonant Silicon Gauge Hot Cathode (Bayard- Alpert) 100 kPa – 100 Pa Direct No Pressure Dependent Change in Resonator Frequency 10 -1 Pa – 10 -7 Pa Indirect Yes Ionization of Gas Molecules Cold Cathode 1 Pa – 10 -7 Pa Indirect Yes Ionization of Gas Molecules “Wide Range” 100 kPa – 10 Pa Indirect Yes, over at least part of the full range. Table 2. Pressure gauges suitable for mass metrology in vacuum. An Introduction to Mass Metrology in Vacuum Patrick J. Abbott, Zeina J. Jabbour Combination of thermal conductivity gauge or piezo resistive gauge with hot or cold cathode gauge Pressure Gauge Recommendations The pressure measurement requirements for mass-invacuum call for a clean, stable gauge to monitor pressure during evacuation and to have less than 10% uncertainty at the pressure of interest where the mass metrology will be performed. In addition to uncertainty constraints, the gauge should not generate any contaminants or an excessive amount of heat. A convection enhanced Pirani gauge is adequate to a pressure of 0.1 Pa, but may be unreliable at lower pressures. In the same way, thermocouple gauges suffer from large uncertainties at their lowest range, around 0.1 Pa. Ionization gauges, both hot cathode and cold cathode, are at their high pressure limits at 0.1 Pa and are greatly influenced by space charge effects. Furthermore, there is evidence that contamination produced by the ionization process in a cold cathode gauge can influence mass measurements [44]. Capacitance manometers, or capacitance diaphragm gauges (CDG) can cover the pressure range between 10 5 to 10 -3 Pa, respond directly to pressure, and when calibrated have an uncertainty of 0.5% or less [45]. Multiple gauge heads are needed to cover the entire range from atmospheric pressure down to 0.1 Pa or less. This can be done using a minimum of two gauge heads having full scale ranges of 133 kPa and 133 Pa. For pressures from atmospheric to 10 Pa, a convection enhanced Pirani gauge will perform sufficiently and may save money over the use of a 133 kPa CDG. Regarding the use of CDG’s, Oct • Nov • Dec 2011 31 Cal Lab: The International Journal of Metrology Cost < ± 1% High < ± 1% Medium ± 0.01% Very High ± 25% Medium ± 25% Medium < 10% above 1 Pa; ± 20% below 1 Pa Medium
Page 1 and 2: METROLOGY 101: PREPARATION & USE OF
Page 3 and 4: www.callabmag.com FEATURES 20 Metro
Page 5 and 6: PUBLISHER MICHAEL L. SCHWARTZ EDITO
Page 7: Best-in-class pressure calibration
Page 10: Jan 24-25 Gage Calibration and Repa
Page 13 and 14: Voltage Transducers Provide an Anal
Page 15 and 16: Huber Petite Fleur - Temperature Co
Page 17 and 18: Mitutoyo CRYSTA-Apex® S CMM The ne
Page 19 and 20: Keithley Launches New Power Supply
Page 22 and 23: Introduction Cal Lab: The Internati
Page 24 and 25: Notes About IEC/ISO17025 Compliance
Page 26 and 27: An Introduction to Mass Metrology i
Page 38 and 39: Effective Communication Between Cus
Page 44: Discover the “Blue Box ” “Blu

Magazine – PDF - Cal Lab Magazine

You also want an ePaper? Increase the reach of your titles

Delete template?

Save as template?